[TowerTalk] Low Inductance Ground Idea

[Jim Lux]

At 09:00 AM 5/13/2004 -0700, Wilson Lui wrote:
>Yes. As long as the metal is non ferrous.
>
>
>
>-----Original Message-----
>From: towertalk-bounces at contesting.com
>[mailto:towertalk-bounces at contesting.com] On Behalf Of GALE STEWARD

<snip>

>  Later, it dawned on me
>that I should have used 3 inch copper instead of the 3
>inch PVC that I installed through the wall to bring my
>coax and control cables into my basement radio room.
>This would provide a low inductance path from the
>radio room to the outside ground system as well as
>providing a raceway for the various cables. I wish I'd
<snip>

Copper pipe is a wonderful material!  Easier to get than copper bar, and 
for RF, lower inductance for a given mass or cost (skin effect and all 
that). If you get a good hot torch, soldering lugs to it, etc. is easy.

Interestingly, there are some recommendations that you'd actually want a 
high inductance/lossy path from outside to inside for the shield, to 
prevent "bad things" on the outside (i.e. lightning induced transients) 
from propagating inside on the outside surface of the coax (or common mode 
on a twisted pair).  One suggestion is to run the cables through a length 
of iron/steel pipe, just for this purpose.

Obviously, if you take a direct hit on the antenna (or a big transient is 
imposed on the antenna from an adjacent hit), you're in trouble, because 
the antenna and feed line is doing just what it's supposed to: carry 
signals from the air to your rig.

However, if you're worried about transients carried in through sneak paths 
(induced on the tower, or on the outside of the coax), then choking the 
outside seems a good idea (somewhat like putting a string of ferrite beads 
over the outside, or clamping a ferrite around the keyboard cable on a 
computer), and iron pipe has a much higher transient handling capability 
than ferrite beads.

One could also do the "shielded room approach" where you bond the coax (all 
the way around, as in a bulkhead connector) to a metal plate in the wall, 
which is then connected to ground by a low impedance connection.  Here, the 
induced potentials on the outside get conducted away just like with any 
shielding scheme.

Something to bear in mind, brought up by Tom a couple weeks ago, is that 
the transients you're trying to protect against might not be coming in 
through your feedline, but through your power line, and if you do a good 
job grounding the rig, then going through your rig might be the best path 
for the transient to go from power line to earth.

I suppose then, the real question is why you'd want a low RF impedance 
ground connection from your rig (inside the house) to earth ground (outside 
the house)?

Presumably, the RF starts out on the "inside" of the box (your rig), and 
stays on the inside of the coax until it gets to the antenna 
feedpoint.  Therefore, the ground from rig to earth shouldn't be carrying 
any RF unless there's some radiated field coupling to the outside.  In this 
case, unless you are also wearing a low impedance ground strap (which is a 
huge safety hazard), you'll be at RF voltage, and the rig will be at zero, 
and there will be a net difference.

Now, if you're driving a long wire antenna with a single wire feedline, 
sure, you need that low impedance RF connection to earth ground, because 
it's the other side of the circuit, however, I suspect most of us use rigs 
with coax connectors on them these days, and the RF ground point (i.e. the 
shield of the coax), such as it is, is at the antenna feedpoint.

It might be true that in a real station, with lots of little boxes, some 
with decent 3rd wire grounds, some without, some with unknown capacitive 
coupling to external sources, that providing a "known good low impedance 
connection" to a common point would be useful. For instance, a lot of 
antenna tuners have no explicit grounding connection other than that 
through the coax, and since no box is perfectly shielding, you could get RF 
on the outside of the coax and the box could be hot relative to the rig.


I've certainly had the unpleasant experience of having racks of equipment 
where the grounding was bad, so one rack floated up to 60-70 VAC relative 
to a "good ground" just from capacitive coupling between AC line and the 
equipment.  This is not a good thing when you have low impedance electrodes 
attached to your head and you touch the rack.

In reality though, solving the "hot chassis" problem with a  "low impedance 
ground strap from equipment to ground point" is a fix for another problem, 
that of imbalance, insufficent shielding, or defective equipment grounding.